ML19210D642

Forwards Response to 790928 Ltr Re 790925 Cooldown Incident. Engineering Analysis Performed Comparing Incident W/Fsar. Detailed Description of Incident & Procedures Used in Response Encl

ML19210D642
Person / Time
Site:	North Anna
Issue date:	11/26/1979
From:	Stallings C VIRGINIA POWER (VIRGINIA ELECTRIC & POWER CO.)
To:	Harold Denton Office of Nuclear Reactor Regulation
References
NUDOCS 7911270420
Download: ML19210D642 (24)

Text

{{#Wiki_filter:*'...VxnorNIA ELECTRIC AND Powun Co31PANY RICHMOND, VIRGIN ZA nO261 Novenber 26, 1979 Mr. Harold R. Denton, Director Serial No. 79 6 Office of Nuclear Reactor Regulation P0/DLB:baw Attn: Mr. A. Schwencer, Chief Docket Nos: 50-338' Operating Reactors Branch #1 50-339 Division of Operating Reactors U. S. Nuclear Regulatory Commission Washington, D. C. 20555'

Subject:

North Anna Unit 1 Cooldown Incident of September 25, 1979 In response to your letter of September 28, 1979, the cttachment provides additional information on the subject incident. Very truly yours, 29. :)M.JaaGf C. M. Stallings Vice President-Power Supply and Production Operations cc: Mr. .Iames P. O'Reilly .1398 300 7911270 iM g W , Y i.*p-, RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION RELATED TO THE SEPTEMBER 25, 1979 EVENT Enclosed are the responses to the infonnation Twmu: 1in yo'Ir September 28, 1979 letter. Should you have any questions or require additional iurm - h . riease contact this office. _1398 301 f..Question 1.Perform an engineering analysis of the Sept. 25, 1979 event, identifying all of the significant initial conditions, and provide a comparison with events analyzed in Chapter 15 of .the North Anna FSAR especially with regard to assumed initial conditions, boundary conditions and system failures. Response: An engineering analysis was performed comparing the Sept. 25, 1979' event with the Chapter 15 events analyzed in the North Anna FSAR. This event most closely follows the accidental depressurization of the Main Steam System analysis - Section 15.2.13 of the North Anna FSAR. Accidental depressurization of the main steam system is classified in the FSAR as a Condition II fault, that is, a fault of moderate frequency. The specific challenge presented by such an event is whether the reactor would, because of the associated primary system cooldown, go critical and experience DNB and possible fuel damage. The case which bounds the consequences of such an event is the spurious opening of a steam dump valve. The nature of the September 25 event was much less severe than the transient shown in the FSAR, since the FSAR analysis includes highly conservative assumptions which were not pre.sent in the September 25 event. Significant differences include the following: 1.Initial TAVG = 570*F, vs. hot no-load condition. This provided additional stored energy, most of which was dissipated by the controlled steam dump operation which brought TAVG to the no-load value.2.Initial power = 78%, vs. no-load in the FSAR. Thus decay heat was available to reduce the cooldown rate and heat the plant after cooldown was terminated. 3.Autmatic feedline isolation, not assumed in the FSAR. This tended to make the rate of cooldown less rapid than in the FSAR. -Manual main steamline isolation, not assumed in the FSAR. This 4.-ended the plant cooldown. 5.Two high-head charging pumps on SI, vs. one (worst single failure assumption) in the FSAR. This, together with the termination of the cooldown and decay heat, induced a rapid repressurization, as opposed to the extended decrease in pressure shown in the FSAR. In both cases safety injection was initiated when the cooldown had adequately reduced pressurizer pressure, and was sufficient to prevent the reactor from returning to a critical condition. Therefore, in both the FSAR -case and the actual incident, DNB was precluded. i Figure i shows the transient response reviewed in the North Anna FSAR and also shows that the plant did not return critical. Therefore DNB did not occur. Figures 2 and 3 show the transient response utilizing the same computer codes and methodology as discussed in the FSAR and Reload Topical (WCAP-9272), however, the assumptions and sequence of events model the Sept. 25, 1979 event at North Anna 1. The transient results show that the pressure reduction, temperature cooldown and margin to criticality are all less limiting than the case analyzed in the FSAR. Thus the FSAR analysis remains bounding. lNO 3.s -n...,, , ,.-2no .If-)!..., mm :n m..e-=, a 8-.'rr-(z n-u.., 1,-;:-(.tj 4: .- 2_1 v.-i .1 e v, , 7. .: r,.;<-c .s <e .1 ,A t---o n.a. m;y er c.t ..-a.: g-(f).o 2 ,..,-u-l<< Ln c:: s:.: r._lw3 i I Ill__}l 5-ig)i-y_-f"';-'&..c_!c ,*.: C r, nv i'o 4 I.T ,.l[.w r.,e v-t ---we!-a[u n .".[v-1.Dr:-a.--<;^ :J. u 4.,5 3 t.._~-,.J--.a e -;-m '"-44 e-a..Q** 1,P*~#*-l.kd!}ll-l1 1~y-<-:-- s g.c-;d a.,u x n. . .. m.. :.x 1 M.--M ,w 0.0'-$:. ." s 1. ' , p s-,.-G. 5 t ~.,.' .; L 4 n."'. f.f3;N.t r , Rl-3j}'.'-n.cc m n1:.us D, 3 6 r, w.+s s: w m. !! .-... '1'.c:--!.3.1.c... h---2 i .- . r,r 3 -h.t,.,.+ ::;.,.(Qx.r!lfllll--2.c^;-.f.. . .....:i- +=..~.s fe n.I ~ . n,, , ,~.~> . c.,.

v. . n . n.,.w. .r . ,n, w, . .~ e .~ . vn O.n n.w s v..o s.. . i y z.eg

.., TIME (3'CO:.:-5} ..., .-l v. . , .@. rf.* - , .; ..

• g /?., 1', . ,f.Fi;;ure 15 2 3 T cm,-~ r .< m.3 r <c.: :, S:r.w ! e B .v t cr .va: :o

.A-Ly$'.'*i- R X2 % v:: at 1C*'O PSm '.'. m %:ww. Nv.t- ^w,ah: $..--3>, I (*#'.- 'D D D'Q..f.Ml2lD M~ . .h[[N a-AQ O-'.S&.q'jh...e-.nj ww w=-3_'-. ;-, 3. -;g.,Y f L 2 , ,,"). :.4**"-;':: ,.--3: i. - C;ps . .,{ tt.--ss.T:.A!;5!E!;T :'Espc ,5E

sp. a c

:

jj-ea e r. y * . . - . . . , . ,' ' % s *. -lwIi e .d.'.*'e , t c.s IN f3;u;n 4 D D'$*t f' l3 OO O:.of.2.A t oo_=..,n ,!-..$. .., (y ,.-4.U[_. l 'uq 2503.0.'~~.s..,.J.j_.;2w.s. r-....s..-- - ---.!..Y ,, T(i. 13 >w-l.,5 7.3 : ,,-X ,,g'.1 v x;--;., w5!.m,g . n.-::2: e a ;-,-. ....m..i j 2.E..s. :.n. .n. A.r-, ,.v ,,, , 4..a.,i, g u e..v;g 150. C3'I-~. "E ,.a g en . W. = _T -,.J.. W '*'e ,; .P I F._'- {d,j-o o-9 O o- O 9 0' j.]'.-o-o e o 0.,#. '-:., ,..1, ]'*m..-c.., b...)' slg . p.' , 3M?*t,,..$ '. 'A u.y 3 .' ' ....s... 3 E .~);',Tg ~%e - n . r =e _e*.*,*_ , .CJv wJ ..e*h .,s y I a.550.03 s 4 1..3#.2 I f i-4,' 'w W*a.p.-.w.J .s..3.....g-i.... .ta 4-.p.aJ. 4.3 4 ..< .4 e-t , p O 8 *J.s 4...**#% M,*1 M ~. ,'-*,, w.o 4.a- 1-.. - :'*NI-M. N --..">o.-u~k , re.u.<#m c , Y....'ti* * " * * " * ' ..-(1 V .; - at . , g3 g *'6-.1j-e=-**r ,-e o o o o ,,.. .!o o o o o o. - . . . . ' . . . . ...... . .-Q o OO O't>O.42-'-o.o o o a..o.. . . : u'8t .. . -*- -... s, , gj.-o-ev M e v)o_ , ...,.%2 -3I. :--..-Ti}-:E (SEC)' .f ;a 1.-t...m w..r-, * :* '<....1T98 304!L)-4. {:sr ---.-a. a :: , . . . .., #b 4.r .dt-,..:6.1-,**.:5 :.: n* t.d'59 l'. 'h & t. . l'n.y^*.E* * *! ,k-[ f. 2 '., l2 r.. . - ~ . . . , c . . n, . , . . V r M , .d. 1. ,~. ., ,. , . -.s.. r..u. _;.., . ,...* ' 'e" A., f. i f.w ' .! .". '.: r'.C. . r ;.". y, p p .C. . i 3 . . ,v. .;.a. .s. ; ..r..1e- - - **..e;--..a ,. . .., j2 ,.t::1'Ij t,.-4.4g<.1 D**T*4 s dc c-..?3 M<-;E--,4 c I.....,-.f..3...h.',[.,. i. -.: I): w-0..-B ,-[-:-.i-c . ., . s., ,t., ,: a..s.g e i..t w.%..u u.T-:. . w ,-.n ,,. , . - -t u.v .--.w . .a -----.-<c.-,..,r-.S...e 5>--= -c..A0;.-y...-, t>. q.:.-j c,...s.., ,. - . - , .7 jg-. ' *u..e..,, 3g: c- . 3..n1.----.w...>= .----- - - - . . > . .-. %,(i b,---s w.a , n_.-c o e o o o a:, ,3 6 m o o o o o e o c, o s,---..., .o o o o o o e e o, 6_o'w: A o o e o a o o--O**c o 8"' 8'O e 4/ 5 40 7.*Cp .b_, M*lo e.9.b, 1,~'.;TIME (SEC)T i*..v ., .y-a.! .:.g.. . , ...,*A . d .8-.l V k.9 4 j. .,'u e.e c...'2 e-4 . ,.!l-I., u>.,.J:,* .-,... ;e-l-. c-y:...._ . _ . .. . . -..... . . . c =.. .: : -9,_42

.....-..._...... . . ... .

'$.+q...v,...... .... ..... .....s .. . . ....&.'--. ., ;* .ge- .--' .. ; ... . . %.1 l'%'.-4,. . '. _.]..y g'. . . .. . .. *. , . . . . .. . . .~.1398 105:. $x r..-.-......... g;-............ --' y.a .'.'J.:),dj.12.$],C,E. &..4.-. c.: 'e.. . - . 4 .Question 2. Identify all plant procedures (number, title, date) used at various phases of the event. Provide a summary describing the extent to which these procedures were used and why this use was considered appropriate. Response 1) EP-1: Reactor Trip (Jan. . 22,1979) - completed entire procedure and is normally used following reactor trips. , 2) EP-3: Main Steam Line Rupture (Sept. 13,1979) - the procedure was only partially completed and was performed because the stuck open steam dump valve simulated a main steam line rupture.

3) AP-41: Safety Injection (Sept. 13, 1979) - the entire

>procedure was completed and performed because the depressurization of the RCS by the open steam dump valve resulted in the actuation of the ECCS.

4) AP-44: Less of RCS Pressure (Sept. 13, 1979) - completed procedure as a result of the RCS depressurization following the failure of the steam dump valve to close.

1396 306.} .. , -- ..3.Describe specifically, the criteria that you used to determine that natural circulation was achieved following RCP trip. Identify the instruments relied upon and their readings. Provide the schedule by which formal written natural circulation cooldown procedures will be available to plant operators. Response: Following the manual trip of the Reactor Coolant Pumps at 0616, the operators ensured that natural circulation was achieved by monitoring the following plant parameters: -A) Constant AT across the core of approximately 30*F as indicated by the Wide Range Cold Leg Temp (480 F) and the Wide Range Hot Leg Temp. (510 F)B) Constant core temperatures as indicated by the Hot and Cold Leg Wide Range Temp. instrumentation. C) Steam Generator Narrow Range Level indication returned and increased which ensured that the S/G tube bundle was covered. D)Increasing Pressurizer Level and Pressure was observed. At approximately 0629, RCP B was restarted which anded the period 2.. at natural: circulation. Formal written natural circulation cooldown procedures will be avail-able to plant operators on or about December 3,1979 which is prior to the restart of the plant. _%1398 307. .Question 4 Provide a description of the reactor coolant pump seal performance during the event. , Response i Prior to the 9/25/79 event, the C RCP at North Anna Unit 1 operated with some variation in leakrate; in general, the trend was low #1 seal leakage and high #2 seal leakage. This leakage variation was reviewed by Westing-house and it was suspected that the #2 seal was not sealing properly. Continued operation was recommenced unless further deterioration was indicated. Although a leakage va;iation existed, the C RCP seals continued to function throughout the 9/25/79 event. Subsequently the C RCP was disassembled and the seals were inspected. The inspection revealed the following: 1)#1 seal assembly - in good condition and was reused. 2)#2 seal insert - some wear and was replaced.

3) #2 seal ring - operating surface worn away.

Replaced>4) #2 seal runner -: operating surface worn away for 45 on it's- circumference. Replaced>5) #3 seal ring - operating surface worn on a taper measuring 0 to 0.034 ' inch.Replaced 6) #3 seal runner - visual appearance was good, but was replaced (for the reasons discussed below). The wear of the #2 ar.d #3 seals was abnormally high. It is suspected that degradation of the aluminum oxide surfaces (of the #2 and #3 seal runners) may have occurred due to high ph levels during hot functional testing; Vepco was advised of this concern via a Westinghouse NSD Technical Bulletin. As a result of this high wear, the #2 seal was not sealing properly allowing more leakage from the #1 seal to pass through the #2 seal and resulting in the indicated low #1 seal leakage and the high #2 seal leakage. The-increased #2 seal leakage was accompanied by increased #3 seal leakage. The hi hi air particulate alarm in the containment attributed to leakage by the

1. 3 seal to the containment atmosphere.

This leakage was observed by operators-to be present only on RCP"C" the day following the event. iW8 308 ..5.Provide an analysis of the actual pressurizer level, including the minimum level reached. , Response: Prior to the reactor trip, the pressurizer level was being maintained at a constant level of 48%. Following the reactor trip and the RCS depressurization, the level began to drop rapidly. At approximately 1 minute prior to safety injection actuation, the level indication went off scale low. Subsequent computer models and mass balance calculations showed that the pressurizer was emptied at about 20 seconds prior to SI. After the initiation of SI, the level began rising rapidly until a maximum level of 73% was achieved and maintained. The pressurizer level began to return to normal when SI was secured and normal letdown and charging were established. 1z98 309. .6.Indicate how many times the PORV cycled and what indications were available to the operator. Explain the apparent 2nd cycling of the PORV approximately 25 mins. af ter the first interval. Response: A PWR cooldown computer model developed by Vepco indicated that the 'PORV cycled approximately 14 times during the first interval and 3 to 4 times during the recond interval. 'The operator estimated between 12 and 24 openings and noted that the valve was not opening fully. Indications to the operator that the PORV was cycling included increased relief line temp; erature; valve position indication lights on the control board; increased temperature, pressure, and level in the pressurizer relief tank; and pressure indication cycling about the setpoint of the PORV on the pressurizer pressure strip chart. The second cycling of the PORV was a result of the pressurizer volume surge, which occurred as a result of letdown being stopped. A sumary of the sequerce of events provides insight into this case. At 0648, about 36 minutes after turbine trip, volume c mcontrol tank high level and high pressure alarms were actuated. As a result, letdown flow was first reduced, and .then temporarily stcpped at 0659. Pressurizer water level and pressure increased and the PORV again was cycled from about 0701 until letdown was restored at 0705. Pressure continued to be sensitive to level changes until about 0820, when the pressurizer heaters finally brought the water in the ;;ressurizer up to saturation temperature. After that time the pressure was held near nonnal while the pressurizer water level was gradually reduced to normal over the next hour. The pressure increase and poor pressure control is attributed to ' lack of normal spray anci pressurizer water subcooling. A similar pressure response would also occur if significant amounts of non-condensible gas were present in the pressurizer. Subsequent analysis of a pressurizer steam space sample, however, did not show unusual non-condensib?e gas concentrations. Therefore, this effect is attributed to lack of normal spray and water subcooling. .1 ,. ..Question 7 Quantify the mass loss through the PORV and explain how this was determined. 'Response: The mass loss through the PORV during the transient was determined to be less than 3500 lb. The method used to calculate this'value was based on assumption that mass loss is directly proportional to the change in pressuri-zer water volume while the PORV was cycling. This assumption is valid since the pressurizer liquid volume was highly subcooled over the duration of the 0 repressurization phase (s 520 F 0 2335 psig). This change in volume can be inferred from the strip charts showing indicated pressurizer level and pressure and hot leg temperature. From this data actual pressurizer level changes can be derived and thus the change in pressurizer volume for both PORVopenjngscanbecalculated. For the first opening a change in volume of 431 Ft occurred which translates into a total steam mass of s3000 lbs. 3 For the second series of PORV openings a volume change of 65 ft occurred which translates into s450 lb. of steam release. A diverse check o' the calcu-lation can be perfonned by using the temperature rise in the Pressurizer Relief Tan!:. The total steam release is assumed to be condensed in the PRT and result in a rise fn PRT temperature. Readings taken at 6 and 10 A.M. on September 25 show a 35 F change in PPT temperature. Assuming a release of 3450 lb. of saturated steam at 2335 psig, the temperature rise would be s51 F.This therefore shows that the calculation is conservative. The primary reason for the deviation is due to the assumption that all steam releases are condensed over this four hour period and that no heat losses are assumed from the PRT walls. .Question 8 Considering the nature of how plant parameters (e.g., pressure, temperature, pressurizer level, etc.) varied and were displayed to the aperator during the event, indicate how and when you would have decided to terminate the high pressure injection based on the HPI termination criteria recomended by Westinghouse. Response: As indicated by the strip chart recordings, wide range reactor coolant temperature T-Hot readings were always greater than 350 F throughout the event, as well as an indicated water level in the steam generators at a level to assure that the U-tubes were covered. Therefore, HPI termination would have relied on satisfying the pressurizer pressure and level' criteria described in Westinghouse procedure E-2. An examination of the transient strip chart recordings and the post accident analysis logs indicates that the pressure criteria (>2000 psia) is satisfied after approximately 0618, or about 4 minutes after safety injection. However, the transient return of indicated pressurizer level during the event was such that the level criterion for HPI termination would not have been satisfied until 0628. At this time all of the Westinghouse requirements for HPI temination would have been satis-fled, and the pressurizer PORV would have subsequently been allowed to close. This is to be compared with the actual pressurizer PORV initial closure time at approximately 0638. Therefore, the initial period during which discharge from the pressurizer PORY occurred would have been approximately 8 minutes, versus that actually recorded during the event of 18 minutes. Since the North Anna event, Westinghouse has proposed revised HPI termina-tion criteria for the E-2 Instruction which modifies the pressurizer level criterien to pennit HPI temination in the absence of abnormal containment indications at a level of 20% span. Based on this criteria, HPI termination during the event could have been accomplished at approximately 0619, thus preventing opening of the pressurizer PORV. 1398 312 t i Question 9 Indicate incore temperature readings taken during the event. Provide details as to magnitude, time and location. Response: Vepco currently has no provision to automatically record incore temperatures following a trip and since the operators were responding to the plant's transient, no incore readings were taken during the natural circula-tion phase of the event. .A computer program which will record incore T/C temperatures after a plant trip is currently being developed. 1398 313 Question 10 Provide the extent to which you consulted with Westinghouse during and immediately following this event. Response: Westinghouse site Systems Engineer was notified of the transient when he reported for work on the morning of the event. The Systems Engineer was requested by VEPCO to contact Westinghouse to perform an evaluation of the reactor vessel cooldown. 1398l4. .Question 11. Provide a detailed chronology of significant events during the period from the initiating event at approximately 0544 hours through return of activity in the auxiliary building to below MPC limits. Response: 0544 LP HX 5B Drain Dump Valve begins cycling 0609 Turbine Trip on High 5B HX Drain Level 0609 Reactor Trip 0610 Aux. FW pumps start on S/G Lo-Lo Level 0610 Pressurizer Pressure 2000 psig -0611 Main FW pump C manually tripped 0611 Letdown Isolation on Low Pressurizer Level 0611 PZR pressure 1885 psig 0611 PZR level 7.4% 0612 Volume Control Tank Low Level Alarm 0613 PZR empties 0614 Safety Injection initiation on Low PZR Pressure 0615 FW Pump A tripped on SI signal 0616 Main steam trip valves manually closed 0616 All RCP's manually tripped 0617 PZR level returned to 9% 0618 PZR pressure 2160 psig 0619 Secured SI signal 0619 Charging pump B secured 0620 PZR PORV starts cycling 0621 FW pump C started 0625 Auxiliary FW secured 0627 Started establishing letdown 0629 Reactor Coolant Pump B started 0631 PZR level 62% 0633 S/G C narrow range level returns on scale 0635 Charging flow realigned & letdown initiated 0637 PZR PORV stops cycling 0646 VCT high pressure and level 69.9 psig and 85.7% 0647 VCT relief valve lifted 0652 First Hi Radiation level alarm - Ventilation Vents 0659 PZR pressure 2320 psig 0659 PORV begins cycling 0704 PORV stops cycling 0716 VCT pressure returns to normal 0730 Aux. Bldg. restricted to access 1000 Aux. Bldg. returned to naimal access 1005 Plant Parameters at normal i M8 315 .Question 12. Yoar alarm typewriter printout indicates that for at least 1 hour prior to the turbine trip, the containment sump level continually cycled to the alarm setpoint. Why? Desponse: The cycling containment sump level is believed to be caused by some leaking check valves downstream of the containment sump pumps. If these valves are not seating correctly, they will pass water through them and back into the sump. This will result in frequent high sump level alanns and near continuous pumping. As a result of the SI signal, the containment sump discharge was isolated. During the event no apparent increase in containment sump level was observed which indicates that there was no leakage of coolant fluids inside the containment prior to or during the transient. The check valves are currently being investigated and if needed will be repaired. ]398 3lI J 4 Question 13. Identify the number of VEPC0 personnel working in the control room during the first 30 minutes of the event. 'Response: At the time of the reactor trip and during the first 30 minutes of the event, the following personnel were present in the control room: the shift supervisor (SR0 license), 2 reactor operators (R0 licenses), and an assistant control room operator trainee. Approximately 15 minutes into the transient the Station Manager arrived in the control room. .1398 317 Question 14. State whether the site emergency plan was activated in any form. Response: Under the guidelines of the Vepco Emergency Plan Implementing Procedure, it was decided not to activate the site emergency plan. This decision was based on knowledge of the implementing procedures which deals with releases of radioactive material. The radioactive releases which occurred following the transient were well below the limits established by 10CFR 20.403 (a) and (b). Also the shift supervisor decided that there was no physical danger to any of the plant workers during or following the transient. Therefore, the decision was made not to implement the site emergency plan. 1798 318> .Question 15: Describe the extent to which the events which occurred at North Anna had been simulated at the Surry Simulator and demonstrated to North Anna operators, prior to Sept. 25, 1979. Discuss the training which North Anna operators have received at the Surry 31mulator on tripping RCP's, natural circulation and Bulletin 79-06B HPI requirements. Response: The combination of the events which occurred at North Anna during the Sept. 25, 1979 cooldown transient had not been simulated at Surry prior to the event. The North Anna cooldown transient was partially simulated at the Surry Simulator after the transient. The problems encountered are as follows: 1.No capability to fail a steam dump open.(All three atmos- .phericdumpswereusedinstead.) 2.Whenever the simulator pressurizer loses level indication, pressure drops to about 1000#, while the actual transient showed pressure staying above 1750#.(Simulate'bynot allowing pressurizer to empty.) 3.Pressure response at the simulator did not follow the transient except for the first two to three minutes. Once pressurizer level increased enough to start cooling the liquid temperature, the pressure indication on the simulator dropped instead of increasing to the PRV setpoint. Pressure stayed low until the pressurizer approached a solid water condition. The PZR appears to be modeled as a saturation vessel with respect to the liquid space temperature. 4.The pressurizer liquid and vapor space temperatures at the simulator did not show the large AT experienced during the actual transient. 5.The release to the Auxiliary Building was not simulated. The cooldown and pressure drop portion of the transient was simulated. Natural circulation indications were approximately the same as experienced on the transient. Once the cause of the excessive cooldown was isolated, the simulator response deviated drastically from the pressure response actually experienced. Due to the simulator response not being like the actual transicnt, there has been no specialized training, on that transient at the Simulator. The transient has been reviewed with the operators currently in training for Unit 2 licenses. The remaining operators have reviewed the LER and will receive additional training during the month of December. p98 M9 2 North Anna operators have received several training sessions devoted to Natural Circulation, tripping RCP's, and procedure changes, HHSI securing criteria, void and hydrogen formation. This training package has been supplied to Operator's Licensing Branch of the NRC as a pre-requisite for Unit 2 licenses. All licensed operators have successfully scored greater than 90% on tests related to this information. All training and testing has been reviewed by OLB and found to be more than satisfactory. 1398 320 .Question 16. Explain in detail the uncontrolled reactor coolant activity release directly to the auxiliary building, a.Include why it happened, how it was detected, its release path (s), how long it continued, the amount, type and form (liquid, gaseous, particulate) of activity released, personnel exposures at the site and potential dose rate at the site boundary and beyond.b.Given the inadvertent operator error, equipment failure, or combination thereof, involved on September 25, 1979, state whether an uncontrolled release would have been prevented had the piping to the process vent from the high level waste drain tanks been installed as called for in the plant design rather than in the as-found conditions. c.If the answer to b above is in the negative, pro-pose a design modification that will prevent a future uncontrolled release of activity outside containment. Response: When normal letdown was established after Safety Injection, the operating charging cump was still aligned to the RWST instead of the VCT. This caused the level in the VCT to rise, at 81% level LCV-1115A diverted all letdown to the gas stripper. High level in the gas stripper system caused the iniet trip valve to the stripper to close. This isolated letdown and letdcwn pressure negan increasing until the low pressure letdown line relief valve coened (200 :sig setpoint). This diverted letdown to the VCT increasing the level and pressure until the VCT relief valve opened (75 psig setpoint) at C647. This initially dumped H2 and radioactive gases to the High Level Liquid Waste Tank (HLLWT). The overpressure condition was not immediately diagnosed and the VCT went water solid and began relieving water to the HLLWT. The HLLWT air sweep to process vent flange for Restricting Orifice LW-104 (RCLW104) was disconnected which allowed the gases to leak directly into the auxiliary building atmosohere. The gases were ultimately discharged to the environment via the plant charcoal and HEPA filters through the auxiliary building ventilation vent. The releases through the auxiliary building ventilation vents provided the first control room indication of Hi Radiation Alarm at 0652. The VCT relief valve centinued cycling until 0704, when the gas stripper was restored and letdown returned to nor:nal. The following information provides the amount, type and times for auxiliary airborne activities. TIME ELEVATION EXPLANATION 0700 274'100.96* Times MPC. Principle Nuclides involved were Xe 133 & 135 with some Kr 85 and RB 88. 259'155.68* Times MPC. Principle Nuclides involved were Xe 133 & 135 with some Kr 85 and RB 88. p98 321 i6-2-TIME ELEVATION EXPLANATION 0800 274'1.12* Times MPC. Principle Nuclides involved were Xe 133 & 135 with traces of Rb 88. 0900 259'6.01* Times MPC. Principle Nuclides involved were Xe 133 & 135 with traces of Rb 88. 1000 259'O.68* Times MPC. Principle Nuclides involved were Xe 133 & 135 with traces of Rb 88. 1030 259'less than 0.1 times MPC. Principle Nuclides involved was Rb 88. All reads after 1030 were less than 0.1 times MPC.

• This value represents the total submersion hazard involved with the total of all isotopes.

Perimeter TLD's were pulled and evaluated. No radiation exposures above background were observed in the 14 TLD's in the downwind direction on the perimeter fence. Total Noble gas releases from ventilation vents A and B and the process vent amounted to 4.7 E-02% of the release rate limit of noble gases. Total personal exposures involved 5 plant workers and the maximum individual dose was 7 mrem. After reviewing the plant design, a release of gases from the VCT to the auxiliary building would not have been prevented had the piping to the process vent from the HLLWT been connected. Calculations show that the volume of gases in the VCT prior to the opening of RV-1257 and the rate of reactor coolant letdown into the VCT during the transient would cause a discharge of gases into the HLLWT at a significantly greater rate than the 6 SCFM air sweep of the process vent system. This condition would result in the gases migrating out the LLLWT overflow vent into the auxiliary building. The sizing of both relief valves was verified to determine their accepta-bility for overpressurization protection. This analysis was done in accordance with ASME Code Case N-94, Determination of Capacities of Liquid Relief Volumes. As discussed above, if RV-1257 opens, the initial surge of high pressure gas will overload the air sweep system and a radioactive release to the auxiliary building would be probable. Since the liquid waste tanks are designed to handle liquid, a design is being evaluated such that the relief path from the VCT be repiped such that liquid would be released instead of gas.By making this modification, the only gas accumulation would be from outgassing as the liquid pressure is reduced from 75 psig to atmospheric pressure. The expected outgassing rate should be within the capacity of the air sweep system. This modification will not impair the overpressurization protection of the VCT because the valve has been designed for liquid service and will pass the required capacity of liquid. 1T98 T22 P Question 17: The FSAR indicates that a high level in the VCT will alann in the control room and divert the letdown stream to the boron recovery system. Did any part of this alarm and diversions system activate prior to or during the time that the VCT relief valve was open? Would such activity affect the release in 16 above?Response: Prior to the event, an RCS dilution of 60 gpm was being maintained to reduce the baron concentration in the RCS. VCT level control valve LCV-lll5A was.in the automatic mode of operation and partially diverting the letdown stream to the gas strippers. When the reactor trip occurred at 0609, the VCT level started decreasing which resulted in LCV-1115A diverting all letdown flow to the VCT.At 0627, operations began to re-establish normal letdown and the VCT level began to rise. By 0646, the VCT level had risen to 86% level at which the letdown flow was being fully diverted to the gas strippers (an 81% of span signal will divert letdown to the stripper). When the letdown flow was diverted to the gas stripper the water level in the stripper began rising to the Hi-Hi Level setpoint of 88% of span. Nonnal action at this time would be an auto start of the stripper discharge pump (1-BR-P-10A) which would transfer the water to the boron recovery tanks. Since the stripper was in manual operation, the pump failed to start and resulted in the stripper inlet valve (TV-BR-111A) tripping closed. At this time letdown flow was blocked which resulted in a pressure increase to the relief valve setpoint limit of 200 psig. Full letdown was diverted to the VCT where pressure increased to the VCT relief valve setpoint of 75 psig. The relief valve opened which relieved water to the High Level Liquid Waste Tank (HLLWT). Meanwhile, the divert valve (LCV-lll5A) remained in the divert to stripper position during the entire time in which the VCT relief valve was opened. With the present design, the activity released to the auxiliary building would have occurred as long as letdown was being maintained with the inlet valve to the gas stripper closed. 1398 323}}